Breast cancer, a genetically heterogeneous disease, is the most common malignancy in women. An estimation of approximately 800000 new cases were reported each year worldwide (Parkin D M, Pisani P, Ferlay J (1999). CA Cancer J Clin 49: 33-64). Mastectomy is the first concurrent option for the treatment of this disease. Despite surgical removal of the primary tumors, relapse at local or distant sites may occur due to undetectable micrometastasis (Saphner T, Tommey D C, Gray R (1996). J Clin Oncol, 14, 2738-2749.) at the time of diagnosis. Cytotoxic agents are usually administered as adjuvant therapy after surgery aiming to kill those residual or pre malignant cells.
Treatment with conventional chemotherapeutic agents is often empirical and is mostly based on histological tumor parameters, and in the absence of specific mechanistic understanding. Target-directed drugs are therefore becoming the bedrock treatment for breast cancer. Tamoxifen and aromatase inhibitors, two representatives of its kind, have been proved to have great responses used as adjuvant or chemoprevention in patients with metastasized breast cancer (Fisher B, Costantino J P, Wickerham D L, Redmond C K, Kavanah M, Cronin W M, Vogel V, Robidoux A, Dimitrov N, Atkins J, Daly M, Wieand S, Tan-Chiu E, Ford L, Wolmark N (1998). J Natl Cancer Inst, 90, 1371-1388; Cuzick J (2002). Lancet 360, 817-824). However the drawback is that only patients expressed estrogen receptors are sensitive to these drugs. A recent concerns were even raised regarding their side effects particularly lay on the possibility of causing endometrial cancer for long term tamoxifen treatment as well as deleterious effect of bone fracture in the postmenopausal women in aromatase prescribed patients (Coleman R E (2004). Oncology. 18 (5 Suppl 3), 16-20). Owing to the emergence of side effect and drug resistance, it is obviously necessarily to search novel molecular targets for selective smart drugs on the basis of characterized mechanisms of action.
Breast cancer is a complex disease associated with numerous genetic changes. Little is known about whether these abnormalities are the cause of breast tumorigenesis, although it has been reported that they occur by a multistep process which can be broadly equated to transformation of normal cells, via the steps of atypical ductal hyperplasia, ductal carcinoma in situ (DCIS) and invasive ductal carcinoma (IDC). There is evidence that only a portion of premalignant lesions are committed to progression to invasive cancer while the other lesions undergo spontaneous regression. This explanation of molecular participation, which leads to development of primary breast cancer, its progression, and its formation of metastases, is the main focus for new strategies targeted at prevention and treatment.
Gene-expression profiles generated by cDNA microarray analysis can provide considerably more detail about the nature of individual cancers than traditional histopathological methods are able to supply. The promise of such information lies in its potential for improving clinical strategies for treating neoplastic diseases and developing novel drugs (Petricoin, E. F., 3rd, Hackett, J. L., Lesko, L. J., Puri, R. K., Gutman, S. I., Chumakov, K., Woodcock, J., Feigal, D. W., Jr., Zoon, K. C., and Sistare, F. D. Medical applications of microarray technologies: a regulatory science perspective. Nat Genet, 32 Suppl: 474-479, 2002). To this aim, the present inventors have analyzed the expression profiles of tumor or tumors from various tissues by cDNA microarrays (Okabe, H. et al., Genome-wide analysis of gene expression in human hepatocellular carcinomas using cDNA microarray: identification of genes involved in viral carcinogenesis and tumor progression. Cancer Res, 61: 2129-2137, 2001; Hasegawa, S. et al., Genome-wide analysis of gene expression in intestinal-type gastric cancers using a complementary DNA microarray representing 23,040 genes. Cancer Res, 62: 7012-7017, 2002; Kaneta, Y. et al., and Ohno, R. Prediction of Sensitivity to ST1571 among Chronic Myeloid Leukemia Patients by Genome-wide cDNA Microarray Analysis. Jpn J Cancer Res, 93: 849-856, 2002; Kaneta, Y. et al., Genome-wide analysis of gene-expression profiles in chronic myeloid leukemia cells using a cDNA microarray. Int J Oncol, 23: 681-691, 2003; Kitahara, O. et al., Alterations of gene expression during colorectal carcinogenesis revealed by cDNA microarrays after laser-capture microdissection of tumor tissues and normal epithelia. Cancer Res, 61: 3544-3549, 2001; Lin, Y. et al. Molecular diagnosis of colorectal tumors by expression profiles of 50 genes expressed differentially in adenomas and carcinomas. Oncogene, 21: 4120-4128, 2002; Nagayama, S. et al., Genome-wide analysis of gene expression in synovial sarcomas using a cDNA microarray. Cancer Res, 62: 5859-5866, 2002; Okutsu, J. et al., Prediction of chemosensitivity for patients with acute myeloid leukemia, according to expression levels of 28 genes selected by genome-wide complementary DNA microarray analysis. Mol Cancer Ther, 1: 1035-1042, 2002; Kikuchi, T. et al., Expression profiles of non-small cell lung cancers on cDNA microarrays: identification of genes for prediction of lymph-node metastasis and sensitivity to anti-cancer drugs. Oncogene, 22: 2192-2205, 2003).
Recent examination into the expression levels of thousands of genes through the use of cDNA microarrays have resulted in the discovery of distinct patterns in different types of breast cancer (Sgroi, D. C. et al., In vivo gene expression profile analysis of human breast cancer progression. Cancer Res, 59: 5656-5661, 1999; Sorlie, T. et al., Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci USA, 98: 10869-10874, 2001; Kauraniemi, P. et al., New amplified and highly expressed genes discovered in the ERBB2 amplicon in breast cancer by cDNA microarrays. Cancer Res, 61: 8235-8240, 2001; Gruvberger, S. et al., S. Estrogen receptor status in breast cancer is associated with remarkably distinct gene expression patterns. Cancer Res, 61: 5979-5984, 2001; Dressman, M. et al., Gene expression profiling detects gene amplification and differentiates tumor types in breast cancer. Cancer Res, 63: 2194-2199, 2003).
Studies into gene-expression profiles in breast cancers have resulted in the identification of genes that may serve as candidates for diagnostic markers or prognosis profiles. However, these data, derived primarily from tumor masses, cannot adequately reflect expressional changes during breast carcinogenesis, because breast cancer cells exist as a solid mass with a highly inflammatory reaction and containing various cellular components. Therefore, previously published microarray data is likely to reflect heterogenous profiles.
Studies designed to reveal mechanisms of carcinogenesis have already facilitated the identification of molecular targets for certain anti-tumor agents. For example, inhibitors of farnesyltransferase (FTIs) which were originally developed to inhibit the growth-signaling pathway related to Ras, whose activation depends on post-translational farnesylation, have been shown to be effective in treating Ras-dependent tumors in animal models (He et al., Cell 99:335-45 (1999)). Similarly, clinical trials on humans using a combination of anti-cancer drugs and the anti-HER2 monoclonal antibody, trastuzumab, with the aim of antagonizing the proto-oncogene receptor HER2/neu have achieved improved clinical response and overall survival of breast-cancer patients (Lin et al., Cancer Res 61:6345-9 (2001)). Finally, a tyrosine kinase inhibitor, STI-571, which selectively inactivates bcr-abl fusion proteins, has been developed to treat chronic myelogenous leukemias wherein constitutive activation of bcr-abl tyrosine kinase plays a crucial role in the transformation of leukocytes. Agents of these kinds are designed to suppress oncogenic activity of specific gene products (Fujita et al., Cancer Res 61:7722-6 (2001)). Accordingly, it is apparent that gene products commonly up-regulated in cancerous cells may serve as potential targets for developing novel anti-cancer agents.
It has been further demonstrated that CD8+ cytotoxic T lymphocytes (CTLs) recognize epitope peptides derived from tumor-associated antigens (TAAs) presented on the MHC Class I molecule, and lyse tumor cells. Since the discovery of the MAGE family as the first example of TAAs, many other TAAs have been discovered using immunological approaches (Boon, Int J Cancer 54: 177-80 (1993); Boon and van der Bruggen, J Exp Med 183: 725-9 (1996); van der Bruggen et al., Science 254: 1643-7 (1991); Brichard et al., J Exp Med 178: 489-95 (1993); Kawakami et al., J Exp Med 180: 347-52 (1994)). Some of the newly discovered TAAs are currently undergoing clinical development as targets of immunotherapy. TAAs discovered so far include MAGE (van der Bruggen et al., Science 254: 1643-7 (1991)), gp100 (Kawakami et al., J Exp Med 180: 347-52 (1994)), SART (Shichijo et al., J Exp Med 187: 277-88 (1998)), and NY-ESO-1 (Chen et al., Proc Natl Acad Sci USA 94: 1914-8 (1997)). On the other hand, gene products demonstrated to be specifically over-expressed in tumor cells have been shown to be recognized as targets inducing cellular immune responses. Such gene products include p53 (Umano et al., Brit J Cancer 84: 1052-7 (2001)), HER2/neu (Tanaka et al., Brit J Cancer 84: 94-9 (2001)), CEA (Nukaya et al., Int J Cancer 80: 92-7 (1999)), and so on.
In spite of significant progress in basic and clinical research concerning TAAs (Rosenberg et al., Nature Med 4: 321-7 (1998); Mukherji et al., Proc Natl Acad Sci USA 92: 8078-82 (1995); Hu et al., Cancer Res 56: 2479-83 (1996)), only limited number of candidate TAAs for the treatment of adenocarcinomas, including colorectal cancer, are currently available. TAAs abundantly expressed in cancer cells yet whose expression is restricted to cancer cells would be promising candidates as immunotherapeutic targets. Further, identification of new TAAs inducing potent and specific antitumor immune responses is expected to encourage clinical use of peptide vaccination strategies for various types of cancer (Boon and can der Bruggen, J Exp Med 183: 725-9 (1996); van der Bruggen et al., Science 254: 1643-7 (1991); Brichard et al., J Exp Med 178: 489-95 (1993); Kawakami et al., J Exp Med 180: 347-52 (1994); Shichijo et al., J Exp Med 187: 277-88 (1998); Chen et al., Proc Natl Acad Sci USA 94: 1914-8 (1997); Harris, J Natl Cancer Inst 88: 1442-5 (1996); Butterfield et al., Cancer Res 59: 3134-42 (1999); Vissers et al., Cancer Res 59: 5554-9 (1999); van der Burg et al., J Immunol 156: 3308-14 (1996); Tanaka et al., Cancer Res 57: 4465-8 (1997); Fujie et al., Int J Cancer 80: 169-72 (1999); Kikuchi et al., Int J Cancer 81: 459-66 (1999); Oiso et al., Int J Cancer 81: 387-94 (1999)).
It has been repeatedly reported that peptide-stimulated peripheral blood mononuclear cells (PBMCs) from certain healthy donors produce significant levels of IFN-γ in response to the peptide, but rarely exert cytotoxicity against tumor cells in an HLA-A24 or -A0201 restricted manner in 51Cr-release assays (Kawano et al., Cancer Res 60: 3550-8 (2000); Nishizaka et al., Cancer Res 60: 4830-7 (2000); Tamura et al., Jpn J Cancer Res 92: 762-7 (2001)). However, both of HLA-A24 and HLA-A0201 are popular HLA alleles in the Japanese, as well as the Caucasian populations (Date et al., Tissue Antigens 47: 93-101 (1996); Kondo et al., J Immunol 155: 4307-12 (1995); Kubo et al., J Immunol 152: 3913-24 (1994); Imanishi et al., Proceeding of the eleventh International Histocompatibility Workshop and Conference Oxford University Press, Oxford, 1065 (1992); Williams et al., Tissue Antigen 49: 129 (1997)). Thus, antigenic peptides of carcinomas presented by these HLAs may be especially useful for the treatment of carcinomas among Japanese and Caucasians. Further, it is known that the induction of low-affinity CTL in vitro usually results from the use of peptide at a high concentration, generating a high level of specific peptide/MHC complexes on antigen presenting cells (APCs), which will effectively activate these CTL (Alexander-Miller et al., Proc Natl Acad Sci USA 93: 4102-7 (1996)).
Accordingly, in an effort to understand the carcinogenic mechanisms associated with cancer and identify potential targets for developing novel anti-cancer agents, the present inventors performed large scale genome-wide analyses of gene expression profiles found in purified populations of breast cancer cells, including 12 ductal carcinomas in situ (DCIS) and 69 invasive ductal carcinomas (IDC), using a cDNA microarray representing 23,040 genes.